Costly Circuits, Submodular Schedules: Hybrid Switch Scheduling for Data Centers

Hybrid switching - in which a high bandwidth circuit switch (optical or wireless) is used in conjunction with a low bandwidth packet switch - is a promising alternative to interconnect servers in today's large scale data-centers. Circuit switches offer a very high link rate, but incur a non-trivial reconfiguration delay which makes their scheduling challenging. In this paper, we demonstrate a lightweight, simple and nearly-optimal scheduling algorithm that trades-off configuration costs with the benefits of reconfiguration that match the traffic demands. The algorithm has strong connections to submodular optimization, has performance at least half that of the optimal schedule and strictly outperforms state of the art in a variety of traffic demand settings. These ideas naturally generalize: we see that indirect routing leads to exponential connectivity; this is another phenomenon of the power of multi hop routing, distinct from the well-known load balancing effects.

[1]  Mounir Hamdi,et al.  On scheduling optical packet switches with reconfiguration delay , 2003, IEEE J. Sel. Areas Commun..

[2]  Yossi Azar,et al.  Efficient Submodular Function Maximization under Linear Packing Constraints , 2012, ICALP.

[3]  Sanjeev Arora,et al.  The Multiplicative Weights Update Method: a Meta-Algorithm and Applications , 2012, Theory Comput..

[4]  William J. Dally,et al.  Guaranteed scheduling for switches with configuration overhead , 2003, IEEE/ACM Trans. Netw..

[5]  Nick McKeown,et al.  Optimal load-balancing , 2005, Proceedings IEEE 24th Annual Joint Conference of the IEEE Computer and Communications Societies..

[6]  Amin Vahdat,et al.  Optics in data center network architecture , 2012 .

[7]  Paolo Giaccone,et al.  Randomized scheduling algorithms for high-aggregate bandwidth switches , 2003, IEEE J. Sel. Areas Commun..

[8]  Ravindra K. Ahuja,et al.  Network Flows: Theory, Algorithms, and Applications , 1993 .

[9]  Chak-Kuen Wong,et al.  Minimizing the Number of Switchings in an SS/TDMA System , 1985, IEEE Trans. Commun..

[10]  Ramin Zabih,et al.  Dynamic Programming and Graph Algorithms in Computer Vision , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Cynthia Barnhart,et al.  A Network-Based Primal-Dual Heuristic for the Solution of Multicommodity Network Flow Problems , 1993, Transp. Sci..

[12]  Srinivasan Seshan,et al.  Scheduling techniques for hybrid circuit/packet networks , 2015, CoNEXT.

[13]  David P. Williamson,et al.  The Design of Approximation Algorithms , 2011 .

[14]  Amin Vahdat,et al.  Integrating microsecond circuit switching into the data center , 2013, SIGCOMM.

[15]  Fillia Makedon,et al.  Fast approximation algorithms for multicommodity flow problems , 1991, STOC '91.

[16]  L. Lovász,et al.  Geometric Algorithms and Combinatorial Optimization , 1981 .

[17]  Kwan Lawrence Yeung,et al.  NXG06-4: Improving Scheduling Efficiency for High-Speed Routers with Optical Switch Fabrics , 2006, IEEE Globecom 2006.

[18]  Xiaohong Jiang,et al.  Cost and delay tradeoff in three-stage switch architecture for data center networks , 2013, 2013 IEEE 14th International Conference on High Performance Switching and Routing (HPSR).

[19]  Fillia Makedon,et al.  Fast Approximation Algorithms for Multicommodity Flow Problems , 1995, J. Comput. Syst. Sci..

[20]  Amin Vahdat,et al.  Helios: a hybrid electrical/optical switch architecture for modular data centers , 2010, SIGCOMM '10.

[21]  Murali S. Kodialam,et al.  Joint Static and Dynamic Traffic Scheduling in Data Center Networks , 2014, IEEE/ACM Transactions on Networking.

[22]  Konstantina Papagiannaki,et al.  c-Through: part-time optics in data centers , 2010, SIGCOMM '10.

[23]  Albert G. Greenberg,et al.  Seawall: Performance Isolation for Cloud Datacenter Networks , 2010, HotCloud.

[24]  He Liu,et al.  Circuit Switching Under the Radar with REACToR , 2014, NSDI.

[25]  Bin Wu,et al.  NXG05-6: Minimum Delay Scheduling in Scalable Hybrid Electronic/Optical Packet Switches , 2006, IEEE Globecom 2006.

[26]  Jean C. Walrand,et al.  Achieving 100% throughput in an input-queued switch , 1996, Proceedings of IEEE INFOCOM '96. Conference on Computer Communications.

[27]  Himanshu Shah,et al.  FireFly , 2014, SIGCOMM.

[28]  Tara Javidi,et al.  Adaptive Policies for Scheduling With Reconfiguration Delay: An End-to-End Solution for All-Optical Data Centers , 2017, IEEE/ACM Transactions on Networking.

[29]  Seth Pettie,et al.  Linear-Time Approximation for Maximum Weight Matching , 2014, JACM.

[30]  Himanshu Shah,et al.  FireFly: a reconfigurable wireless data center fabric using free-space optics , 2015, SIGCOMM 2015.

[31]  P. Sanders,et al.  A simpler linear time 2 / 3 − ε approximation for maximum weight matching , 2004 .

[32]  Nick McKeown,et al.  A practical scheduling algorithm to achieve 100% throughput in input-queued switches , 1998, Proceedings. IEEE INFOCOM '98, the Conference on Computer Communications. Seventeenth Annual Joint Conference of the IEEE Computer and Communications Societies. Gateway to the 21st Century (Cat. No.98.

[33]  Ben Y. Zhao,et al.  Mirror mirror on the ceiling: flexible wireless links for data centers , 2012, SIGCOMM.

[34]  J. G. Pierce,et al.  Geometric Algorithms and Combinatorial Optimization , 2016 .

[35]  Robert E. Tarjan,et al.  Fibonacci heaps and their uses in improved network optimization algorithms , 1984, JACM.

[36]  Albert G. Greenberg,et al.  Towards a next generation data center architecture: scalability and commoditization , 2008, PRESTO '08.

[37]  Ankit Singla,et al.  OSA: An Optical Switching Architecture for Data Center Networks With Unprecedented Flexibility , 2012, IEEE/ACM Transactions on Networking.

[38]  Lei Ying,et al.  Communication Networks - An Optimization, Control, and Stochastic Networks Perspective , 2014 .

[39]  R. Srikant,et al.  Optimal WDM schedules for optical star networks , 1999, TNET.

[40]  Cheng-Shang Chang,et al.  Birkhoff-von Neumann input buffered crossbar switches , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).

[41]  Nick McKeown,et al.  On the speedup required for combined input- and output-queued switching , 1999, Autom..

[42]  Murali S. Kodialam,et al.  On guaranteed smooth scheduling for input-queued switches , 2005, IEEE/ACM Transactions on Networking.

[43]  Alex C. Snoeren,et al.  Inside the Social Network's (Datacenter) Network , 2015, Comput. Commun. Rev..

[44]  Michael O. Rabin,et al.  Efficient dispersal of information for security, load balancing, and fault tolerance , 1989, JACM.

[45]  Leslie G. Valiant,et al.  A bridging model for parallel computation , 1990, CACM.

[46]  Paramvir Bahl,et al.  Flyways To De-Congest Data Center Networks , 2009, HotNets.

[47]  Shivendra S. Panwar,et al.  Frame-based matching algorithms for optical switches , 2003, Workshop on High Performance Switching and Routing, 2003, HPSR..

[48]  T. Inukai,et al.  An Efficient SS/TDMA Time Slot Assignment Algorithm , 1979, IEEE Trans. Commun..

[49]  Nick McKeown,et al.  The iSLIP scheduling algorithm for input-queued switches , 1999, TNET.

[50]  Tara Javidi,et al.  End-to-end scheduling for all-optical data centers , 2015, 2015 IEEE Conference on Computer Communications (INFOCOM).

[51]  Albert G. Greenberg,et al.  Data center TCP (DCTCP) , 2010, SIGCOMM '10.

[52]  David A. Maltz,et al.  Network traffic characteristics of data centers in the wild , 2010, IMC '10.

[53]  Alberto Leon-Garcia,et al.  Communication Networks , 2000 .

[54]  N. Biggs.  GEOMETRIC ALGORITHMS AND COMBINATORIAL OPTIMIZATION: (Algorithms and Combinatorics 2) , 1990 .

[55]  Jochen Könemann,et al.  Faster and simpler algorithms for multicommodity flow and other fractional packing problems , 1998, Proceedings 39th Annual Symposium on Foundations of Computer Science (Cat. No.98CB36280).

[56]  William J. Dally,et al.  Guaranteed scheduling for switches with configuration overhead , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[57]  Jochen Könemann,et al.  Faster and Simpler Algorithms for Multicommodity Flow and Other Fractional Packing Problems , 2007, SIAM J. Comput..

[58]  Alexander Schrijver,et al.  Combinatorial optimization. Polyhedra and efficiency. , 2003 .

[59]  Cheng-Shang Chang,et al.  Load balanced Birkhoff-von Neumann switches , 2001, 2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552).

[60]  Ben Y. Zhao,et al.  Mirror mirror on the ceiling: flexible wireless links for data centers , 2012, CCRV.

[61]  Ran Duan,et al.  A scaling algorithm for maximum weight matching in bipartite graphs , 2012, SODA.